Deep learning for fake news detection: A comprehensive survey

Hu, Linmei; Wei, Siqi; Zhao, Ziwang; Wu, Bin

doi:10.1016/j.aiopen.2022.09.001

Cited by 68 publications

(26 citation statements)

References 125 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, the latest Weibo annual report on fake news [38] revealed that Weibo's official fact-checking agency identified 82,274 pieces of fake news in 2022. Given the devastating consequences of fake news on both individuals and society, fake news detection has become an urgent and essential task that needs to be addressed [1,9,18,[26][27][28]36]. To this end, Chinese fake news detection datasets have been constructed for the development of Chinese fake news detection [9-12, 18, 40, 41].…”

Section: Introductionmentioning

confidence: 99%

“…However, in the real world, news emerges from multiple sources, such as social platforms, messaging apps, and traditional online news outlets, etc. [3,9,13,14,20]. News, in particular, fake news, from different sources is characterized by diverse dimensions, such as news content, topics, publishing methods, and the utilization of sophisticated linguistic styles intended to mimic real news [1,19,21,31,37].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MCFEND: A Multi-source Benchmark Dataset for Chinese Fake News Detection

Li,

He,

Bai

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

The prevalence of fake news across various online sources has had a significant influence on the public. Existing Chinese fake news detection datasets are limited to news sourced solely from Weibo. However, fake news originating from multiple sources exhibits diversity in various aspects, including its content and social context. Methods trained on purely one single news source can hardly be applicable to real-world scenarios. Our pilot experiment demonstrates that the F1 score of the state-of-the-art method that learns from a large Chinese fake news detection dataset, Weibo-21, drops significantly from 0.943 to 0.470 when the test data is changed to multi-source news data, failing to identify more than one-third of the multi-source fake news. To address this limitation, we constructed the first multi-source benchmark dataset for Chinese fake news detection, termed MCFEND, which is composed of news we collected from diverse sources such as social platforms, messaging apps, and traditional online news outlets. Notably, such news has been fact-checked by 14 authoritative fact-checking agencies worldwide. In addition, various existing Chinese fake news detection methods are thoroughly evaluated on our proposed dataset in cross-source, multi-source, and unseen source ways. MCFEND, as a benchmark dataset, aims to advance Chinese fake news detection approaches in real-world scenarios.Haorui He and Dacheng Wen are also with

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

MCFEND: A Multi-source Benchmark Dataset for Chinese Fake News Detection

Li,

He,

Bai

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

show abstract

“…Computational approaches focus on the definition of the misinformation concept [33], its propagation mechanisms [13,60], and detection and preventing its propagation [34,44,55,65]. Recent studies on this field iterate over various types of textual, behavioral, and media aspects to identify misinformation using machine learning as explained by [28], [70], and [68]. Computational efforts should also take the concepts of decentralization, transparency, and objectivity into account and should have demonstrative aspects against tampering for any outside agenda [38,40,51].…”

Section: Introductionmentioning

confidence: 99%

Modeling and Mitigating Online Misinformation: a Suggested Blockchain Approach

Yılmaz¹,

Ulusoy²

2023

Preprint

View full text Add to dashboard Cite

Misinformation propagation in online social networks has become an increasingly challenging problem. Although many studies exist to solve the problem computationally, a permanent and robust solution is yet to be discovered. In this study, we propose and demonstrate the effectiveness of a blockchain-machine learning hybrid approach for addressing the issue of misinformation in a crowdsourced environment. First, we motivate the use of blockchain for this problem by finding the crucial parts contributing to the dissemination of misinformation and how blockchain can be useful, respectively. Second, we propose a method that combines the wisdom of the crowd with a behavioral classifier to classify the news stories in terms of their truthfulness while reducing the effects of the actions performed by malicious users. We conduct experiments and simulations under different scenarios and attacks to assess the performance of this approach. Finally, we provide a case study involving a comparison with an existing approach using Twitter Birdwatch data. Our results suggest that this solution holds promise and warrants further investigation. CCS Concepts: • Computing methodologies → Neural networks; • Computer systems organization → Dependable and fault-tolerant systems and networks; • Information systems → Crowdsourcing.

show abstract

“…There are a number of existing surveys on fake news detection. However, the majority of works (e.g., [2]- [5]) focus on providing an overview of the entire field of fake news detection, and graph-based deep learning methods are either not mentioned at all [2] or only briefly described [3]- [5]. This lack of coverage creates a mismatch given the recent surge in the use of graph-based methods for fake news detection and the impressive results they achieve.…”

Section: Introductionmentioning

confidence: 99%